Atmospheric Environment 42 (2008) 9106–9117
Contents lists available at ScienceDirect
Atmospheric Environment
journal homepage: www.elsevier.com/locate/atmosenv
Seasonality of the Na/Cl ratio in precipitation and implications
of canopy leaching in validating chemical analyses
of throughfall samples
Anne Thimonier a, *, Maria Schmitt a, Peter Waldner a, Patrick Schleppi a
a
WSL, Swiss Federal Institute for Forest, Snow and Landscape Research, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 June 2008
Received in revised form 28 August 2008
Accepted 3 September 2008
The temporal variation of the Na/Cl ratio in bulk precipitation and throughfall samples was
analysed using an extensive data set based on measurements over several years at 11 sites
of the Swiss Long-Term Forest Ecosystem Research Programme (LWF). This analysis was
prompted by the results of the application of the criteria recommended for validating
chemical analyses of precipitation samples in the Integrated Co-operative Programme on
Assessment and Monitoring of Air Pollution Effects on Forests (ICP Forests). One of these
criteria involves verifying that the sodium to chloride (Na/Cl) ratio of the analysed sample
is within a restricted range, assuming that most of the chloride originates from sea-spray,
and that the contribution of marine ions in atmospheric deposition is in the same molar
ratio (0.86) as in sea-water. The range of possible Na/Cl values was defined between 0.5
and 1.5 by ICP Forests in order to account for other possible sources of Naþ and Cl (natural
or anthropogenic). When all sites were considered, approx. 85% of our Na/Cl values were
within the proposed range, both for bulk precipitation and throughfall samples. In some
cases, low Naþ or Cl concentrations close to the detection limit were responsible for the
Na/Cl ratios outside the range of acceptance. Plotting the Na/Cl ratio versus time revealed
a seasonal pattern, which was clearer in the throughfall than in the bulk precipitation
samples. This could also account for Na/Cl values higher or lower than the defined limits.
The seasonality of the Na/Cl ratio and its components (Naþ and Cl fluxes) was tested using
a regression model. For throughfall, the seasonal pattern of Na/Cl could be ascribed to the
seasonally driven canopy leaching of Naþ and Cl, the intensity of which depended on the
tree species.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Na/Cl ratio
Precipitation chemistry
Canopy leaching
Seasonal variation
Quality assurance
1. Introduction
High quality measurements of atmospheric deposition
are required in several areas of ecological research such as
studies of air quality, water quality or ecosystems. Total
atmospheric deposition includes a remarkable amount of
dry deposition, especially in forests. Dry deposition can be
measured by the inferential method, the micrometeorological
* Corresponding author. Tel.: þ41 44 739 23 55; fax: þ41 44 739 22 15.
E-mail address: [email protected] (A. Thimonier).
1352-2310/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.atmosenv.2008.09.007
method (Wesely and Hicks, 2000), or the throughfall
method. While the first two methods require highly
sophisticated equipment, the latter relies on relatively
simple measurements but implies assumptions about the
exchange of ions in the canopy. In particular, negligible
uptake and leaching are assumed for sodium (Naþ) and
chloride (Cl).
In Europe, the atmospheric deposition of nutrients and
pollutants in forests is one of the key issues addressed by
the Integrated Co-operative Programme on Assessment
and Monitoring of Air Pollution Effects on Forests (ICP
Forests), which aims at a better understanding of how
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
9107
Table 1
Site characteristics. Mean annual precipitation amounts were derived from precipitation maps established for the period 1961–1990 (FOWG, 2000). Snow
cover after Witmer (1986).
Region
Longitude
Latitude
Altitude
a.s.l.
(m)
Aspect
Mean
slope
(%)
Annual
precipitation
(mm)
Number of
days with
snow cover
Main tree species
07 250 E
47 140 N
1149
S
66
1549
120
Fagus sylvatica L.
06 170 E
06 400 E
08 140 E
07 530 E
46 140 N
46 350 N
47 240 N
47 170 N
501
807
484
480
flat
NE
S
NW
3
7
27
14
977
1210
1123
1115
25
63
57
51
09 040 E
07460 E
47 100 N
46 430 N
733
1511
W
SW
60
33
1801
1305
97
133
Fagus sylvatica L.
Picea abies (L.) H.Karst.
09 530 E
10 140 E
07 520 E
46 300 N
46 400 N
46 180 N
1871
1899
695
NE
S
N
34
11
80
1024
1048
689
172
189
55
Pinus cembra L.
Pinus mugo Turra
Pinus sylvestris L.
08 500 E
46 010 N
950
S
68
2017
48
Quercus cerris L.
Site
Jura
Bettlachstock
Central Plateau
Jussy
Lausanne
Othmarsingen
Vordemwald
Lower Alps
Schänis
Beatenberg
Alps
Celerina
National Park
Visp
Southern Alps
Novaggio
natural and anthropogenic stresses affect forest ecosystems
in the long term (de Vries et al., 2003). Atmospheric
deposition is currently measured in over 500 ICP Forests
plots across Europe using the throughfall method. This
method involves the parallel sampling of precipitation
below the forest canopy (throughfall) and in an open area
near the forest stand, using either continuously open
collectors (bulk deposition) or wet-only collectors (wet
deposition). Deposition measurements have been carried
out for over a decade now, but quality control has always
been done at the national level. It was only a few years ago
that the problem of the quality and thus the comparability
of the chemical analyses was addressed at the European
level. For this purpose, a working group on quality assurance and quality control (QA/QC) was created within the
Expert Panel on Deposition (Mosello et al., 2005). This
Expert Panel, which is part of ICP Forests, includes national
experts from all participating countries. It ensures the
development and harmonization of the monitoring
methods and contributes to data evaluations in the field of
atmospheric deposition.
3
1998
1999
2000
2001
Quercus robur L., Carpinus betulus L.
Fagus sylvatica L.
Fagus sylvatica L.
Abies alba Mill.
The Working Group on QA/QC has made (and is still
making) an important contribution to the improvement of
the overall quality of the measurements by addressing
quality aspects not only in the field and the laboratory (e.g.
evaluation of the analytical methods) but also in the validation of results. In the validation stage, four consistency
checks on chemical data are currently recommended (Mosello
et al., 2005): 1) the ion balance, 2) a comparison between
measured and calculated conductivities, 3) a comparison
between the sum of the inorganic forms of nitrogen and the
total nitrogen, and 4) the Na/Cl ratio. In this paper we focus
on the Na/Cl ratio.
In areas close to the sea, sea salt is a major contributor to
sodium and chloride deposition, and the molar Na/Cl ratio
in precipitation is typically that of sea salt, namely 0.86
(Keene et al., 1986). However, the calculation of the Na/Cl
ratio has also proved useful in sites far from the sea. The ICP
Forests Manual proposed a range between 0.5 and 1.5
(molar values) as being acceptable (Ulrich et al., 2006).
Samples with a Na/Cl ratio outside this range should be reanalyzed. However, if the second run of analyses confirms
2002
2003
2004
2005
2006
2007
Na/CI
2
1
0
J A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ
Fig. 1. Temporal variation of the molar Na/Cl ratio in bulk precipitation (dots) at Jussy, and regression curve (bold line) calculated with data after 01.01.2001
(Na/Cl ¼ 1.098 þ 0.05 cos t 0.1 sin t 0.06 cos 2t þ 0.113 sin 4t; R2 ¼ 0.17, P < 0.0001).
9108
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
Table 2
Variability of the Na/Cl ratio with concentrations (c) close to the determination limit (L). The uncertainty D of a measured concentration is defined as D ¼ max
(L, 0.10 c). Example with cNa ¼ 2 LNa and Na/Cl ¼ 1.
Naþ
Cl
Concentration,
c
[mg L1]
Determination limit,
L
[mg L1]
D
0.16
0.25
0.08
0.01
Uncertainty,
[mg L1]
Lower limit,
cD
[mg L1]
Upper limit,
cþD
[mg L1]
Concentration,
c
[meq L1]
Lower concentration,
cD
[meq L1]
Upper concentration,
cþD
[meq L1]
0.08
0.025
0.08
0.225
0.24
0.275
6.96
7.04
Ratio
1.0
3.48
6.34
Lower ratioa
0.4
10.43
7.75
Upper ratiob
1.6
Na/Cl
a
b
The lower Na/Cl ratio is calculated as the ratio between the lower Naþ concentration and the upper Cl concentration.
The upper Na/Cl ratio is calculated as the ratio between the upper Naþ concentration and the lower Cl concentration.
the concentrations obtained initially, the analyses may then
be validated even when the thresholds are exceeded.
The validation procedures recommended by the ICP
Forests Manual have been routinely applied for some
years in the deposition monitoring activities of the Swiss
Long-Term Forest Ecosystem Research programme (LWF;
Cherubini and Innes, 2000; Thimonier et al., 2001).
However, the occurrence of repeated deviations from the
acceptable range under certain conditions, especially in
throughfall samples, led us to assess the applicability of
the Na/Cl ratio check more closely.
2. Material and methods
2.1. Sites
Bulk precipitation and throughfall are currently regularly collected at 11 LWF sites distributed across the five
main regions of Switzerland (Table 1; Thimonier et al.,
2005). On the Central Plateau, which is the most densely
populated region in Switzerland, the deposition levels are
moderate. Deposition rates decrease with increasing altitude and are lowest in the Alps. They are highest (up to
35 kg ha1 a1 for N and 17 kg ha1 a1 for S in throughfall)
in Southern Switzerland, which is subjected to the emissions from the industrialised and densely populated Po
Basin (Thimonier et al., 2005). In this paper, we will focus
on one site in particular, Jussy, where deviations of Na/Cl
from the acceptable range have been most frequent. The
site of Jussy is located near Geneva on the Central Plateau.
On this site, the forest stand is dominated by oak (Quercus
robur L. and Quercus petraea (Matt.) Liebl.) and hornbeam
(Carpinus betulus L.). We will also present detailed results
for the site of Celerina, a high-elevation site in the Alps,
with Swiss stone pine (Pinus cembra L.) as the main tree
species.
2.2. Sampling procedures
A detailed description of the sampling procedures is
given in Thimonier et al. (2005). Bulk precipitation was
collected with three funnel-type polyethylene collectors
(100 cm2 opening). In winter at the sites where abundant
snowfall can be expected (Bettlachstock, Beatenberg,
Schänis, Celerina, National Park, Lausanne and Novaggio),
the funnel-type collectors were replaced by a single
bucket-type snow collector (30 cm diameter).
Throughfall was sampled with 16 funnel-type collectors of the same design as the collectors used in the open
area. The collectors were systematically distributed over
two 43 43 m subplots. In winter at the sites where
abundant snowfall was expected (see above), the 16
funnel-type collectors were replaced by four bucket-type
collectors.
The collecting samplers were collected once every two
weeks (four weeks at Celerina and Bettlachstock in the
winter), and replaced by new ones. All samples were sent
by post or brought directly to the WSL Research Institute at
Birmensdorf, which coordinates the monitoring activities
and evaluates the data. Within 3 days of arrival, the
samples were filtered (0.45 mm) and the conductivity and
the pH were measured in the laboratory. All samples were
prepared in duplicate, one for the chemical analyses of the
macro-elements by the WSL central laboratory, the other
for storage at þ2 C to allow for repetition of the analyses
should the validation checks reveal inconsistencies.
Table 3
Seasonal variation of Na/Cl in bulk precipitation at 11 LWF sites, sorted according to the geographic region. The R2 values are the coefficients of determination
of the regression models for the seasonality. The unweighted median Na/Cl ratio, the lower quartile (P25) and the upper quartile (P75) are reported for
each site.
Region
Site
n
Median
P25
P75
R2
P
Peaks
Jura
Central Plateau
Bettlachstock
Jussy
Lausanne
Othmarsingen
Vordemwald
Beatenberg
Schänis
Celerina
National Park
Visp
Novaggio
126
158
158
165
166
162
163
118
157
120
145
1.03
1.05
1.14
1.06
1.04
1.18
1.14
0.95
0.88
1.00
1.07
0.88
0.90
0.99
0.91
0.89
0.96
0.99
0.72
0.73
0.85
0.90
1.29
1.25
1.39
1.23
1.19
1.41
1.54
1.18
1.13
1.23
1.28
0.12
0.17
0.10
0.02
0.02
0.13
0.14
0.17
0.20
0.10
0.11
P < 0.001
P < 0.0001
P < 0.01
n.s.
n.s.
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.01
P < 0.01
Summer
Autumn
Summer
Lower Alps
Alps
Southern Alps
Summer
Summer
Late summer
Late summer
Spring, autumn
Lowest in winter
3.1. Bulk precipitation
3.1.1. Seasonal variation of the Na/Cl ratio
Fig. 1 illustrates the temporal variation of the Na/Cl ratio
in bulk precipitation at the LWF site of Jussy from 1998 to
mid 2007. It shows a wider spread of the values before
End of winter
P < 0.0001
0.15
59.1
8.9
End of winter
P < 0.001
0.12
59.9
9.9
Spring to autumn
0.17
0.07
0.04
0.13
118
160
120
145
P < 0.001
End of winter
Winter
End of winter
P < 0.0001
P < 0.0001
P < 0.01
0.21
0.12
0.11
22.8
33.0
124.4
5.5
6.4
12.7
End of winter
Winter
End of winter
P < 0.01
P < 0.001
P < 0.01
0.10
0.08
0.12
21.1
29.6
122.7
5.5
6.1
12.3
Spring to autumn
Spring to autumn
0.28
0.16
162
163
P < 0.001
P < 0.01
P < 0.05
Winter
End of winter
P < 0.01
P < 0.05
0.06
0.04
34.6
72.2
4.3
6.4
Winter
End of winter
P < 0.05
P < 0.05
0.05
0.07
31.7
67.1
5.0
7.5
Spring to autumn
Spring to autumn
Spring to autumn
Spring to autumn
P < 0.001
P < 0.05
n.s.
0.02
0
0.11
0.04
158
158
164
166
P < 0.0001
P < 0.0001
End of winter (early spring)
Winter
Winter (summer)
Winter (summer)
P < 0.05
P < 0.0001
P < 0.0001
P < 0.01
0.03
0.16
0.15
0.09
40.9
42.3
40.8
32.6
7.8
7.2
7.4
7.6
End of winter
Winter
Winter (summer)
Winter (summer)
P < 0.05
P < 0.0001
P < 0.0001
P < 0.001
0.05
0.14
0.19
0.12
32.9
41.6
39.8
31.0
8.2
8.2
7.8
7.8
Winter
P < 0.0001
0.19
37.8
6.3
Winter
P < 0.0001
0.17
33.0
6.8
Spring to autumn
Peaks
P
R2
s.d.
Mean
Mean
s.d.
R2
P
Peaks
Cl concentrations [meq L1]
Peaks
P
P < 0.05
0.07
127
3. Results
R2
where t ¼ 2p/365 $ (day of the year) and c1–c11 are
regression coefficients.
We included sine and cosine terms up to 5t in order to
best reproduce the systematic narrow peaks clearly visible
on the measured data set. Sine and cosine terms with 5t
correspond to a period of the sinusoidal curve of 365/5 ¼ 73
days, which allows the model to show processes which go
up or down within a little more than one month. This
makes the analysis sensitive enough to reveal the important features of the annual cycle, but not too sensitive to
single irregularities.
When analysing the seasonality of throughfall fluxes,
we applied the regression model to the difference between
throughfall and bulk precipitation (net throughfall, also
called enrichment) rather than directly to the throughfall
fluxes. This allowed us to remove the variability due to
variations in incident precipitation.
n
þ c10 sin 5t þ c11 cos 5t
Site
þ c6 sin 3t þ c7 cos 3t þ c8 sin 4t þ c9 cos 4t
9109
Jura
Bettlachstock
Central Plateau
Jussy
Lausanne
Othmarsingen
Vordemwald
Lower Alps
Beatenberg
Schänis
Alps
Celerina
National Park
Visp
Southern Alps
Novaggio
x ¼ c1 þ c2 sin t þ c3 cos t þ c4 sin 2t þ c5 cos 2t
Naþconcentrations [meq L1]
The seasonality within the data was tested by introducing harmonic terms (sine and cosine) in a stepwise
regression of the dependent variable x versus time (e.g.
Schleppi et al., 2006), in the form:
Precipitation [mm]
2.3. Data analysis
Region
Ammonium (NHþ
4 ) was determined colorimetrically
through automated flow injection analysis. Calcium (Ca2þ),
magnesium (Mg2þ), potassium (Kþ) and sodium (Naþ)
were determined by inductively coupled plasma–atomic
2
emission spectrometry. Nitrate (NO
3 ), sulphate (SO4 ) and
chloride (Cl) concentrations were analysed by ion chromatography. Determination limits for Naþ and Cl were
0.08 mg L1 and 0.01 mg L1, respectively. Dissolved organic
carbon (DOC) and total dissolved nitrogen have been analysed using a TOC-V analyser (Shimadzu, Tokyo, Japan)
since May 2001. Dissolved organic nitrogen (DON) was
calculated as the difference between total nitrogen and
inorganic nitrogen (NHþ
4 –N þ NO3 –N). All the analyses
were checked against certified standards. International
comparison exercises (e.g. Marchetto et al., 2006)
confirmed that the analyses carried out by the WSL laboratory were satisfactory for all ions. One exception was total
dissolved nitrogen, which tended to be underestimated at
high nitrogen concentrations. The ion balance, the conductivity derived from the ionic composition of the samples
and the Na/Cl molar ratio, were calculated for each sample
as soon as all the chemical analyses were completed.
Table 4
Seasonal variation of bulk precipitation volumes and Naþ and Cl concentrations. Average Naþ and Cl concentrations are volume weighted. The R2 values are the coefficients of determination of the regression
models for the seasonality.
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
9110
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
R2 of 0.17, significant at the P < 0.0001 level (Fig. 1, Table 3).
Na/Cl in bulk precipitation tended to be higher in the
autumn at Jussy.
Seasonal patterns of the Na/Cl ratio in bulk precipitation
were also observed at the two high-elevation sites (above
1800 m) in the Alps, at Celerina and the Swiss National
Park, with higher Na/Cl ratios being measured in the late
summer (Table 3). Lower, but still highly significant coefficients of determination were obtained at mid-elevation
sites in the Jura mountains (Bettlachstock) and in the Lower
Alps, with higher Na/Cl ratios in the summer. Jussy was the
site on the Plateau where R2 was the highest. There was no
seasonality at the sites in Othmarsingen and Vordemwald,
which are both below 500 m. The median Na/Cl ratio in
bulk precipitation was higher (ranging from 0.88 in the
National Park to 1.18 at Beatenberg) than the ratio in seawater on all sites. If all sites were considered, 84% of all Na/
Cl values were within the range of acceptance (0.5–1.5).
2000–2001, with several values above and below the limits
of the acceptable range. Before 2001, the Na/Cl criteria was
not systematically used in the validation step, and occasional contamination with Naþ or Cl due to inappropriate
procedures cannot be excluded. Naþ and Cl contaminations usually occurred independently of each other, as
these two elements are analysed with different analytical
methods and instruments. Washed polypropylene tubes
are used for Naþ, and new, unwashed glass chromatography vials for Cl.
Since 2001, several samples have had a Na/Cl ratio
higher than 1.5. A closer look at these samples with
extreme ratios revealed that they have low concentrations
of Naþ and Cl, including the sample with the highest ratio
observed in 2003. For such samples, the calculated ratio is
very sensitive to very small deviations in the concentrations. The quality assurance procedure for the chemical
analyses in our laboratory are designed to ensure that the
uncertainty D of the measured concentration c can be described with D ¼ max(L, 0.10 c), where L is the determination
limit. Table 2 illustrates with an example how the uncertainty about concentrations close to the determination
limit also effects the uncertainty on the Na/Cl ratio: with
a measured concentration cNa ¼ 2LNa and Na/Cl y 1,
uncertainties about Naþ and Cl concentrations result in
Na/Cl values ranging from 0.4 to 1.6. In such cases it thus
seems appropriate to extend the range of acceptable values
for the Na/Cl ratio.
In order to remove the influence of possible Naþ or Cl
contaminations prior to 2001, we tested the seasonality
only in the data collected after January 1st, 2001.
The regression model introducing sine and cosine
functions of time resulted in a coefficient of determination
3.1.2. Seasonal variation of Na and Cl concentrations and fluxes
The highest concentrations of Naþ and Cl in bulk
precipitation were usually measured in winter (Table 4).
This was often related to the seasonal variation in precipitation volumes, which tended to be significantly lower in
winter than during the rest of the year.
The site of Celerina, where the seasonal variation of Na/
Cl was highly significant, was taken as an example representative of the other sites regarding precipitation and
concentration patterns. Fig. 2 illustrates the parallel variations of Naþ and Cl concentrations and fluxes on this
particular site.
At Celerina, concentrations of Naþ and Cl were highest
at the end of the winter. At this time of year, Cl
50
1.2
1.1
Na/Cl ratio [-] and Ion flux
[meq m-2 (2 weeks)-1]
40
0.9
0.8
30
0.7
0.6
0.5
20
0.4
0.3
10
0.2
Water flux [mm (2 weeks)-1] and
Ion concentration [µeq L-1]
1.0
0.1
0
0.0
J
F
M
A
M
-
[Cl ]
-
Cl flux
J
J
A
S
O
[Na+]
Na/Cl
Na+ flux
Water flux
N
D
Fig. 2. Modelled seasonal variation of Na/Cl, Naþ and Cl concentrations and fluxes, and bulk precipitation volume at Celerina (concentrations, fluxes and ratio:
data after 01/01/2001; precipitation volume: all data since the beginning of the measurements, 13/07/1999).
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
3
1998
1999
2000
2001
2002
2003
2004
9111
2005
2006
2007
Na/CI
2
1
0
J A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ
Fig. 3. Temporal variation of the molar Na/Cl ratio in throughfall (dots) at Jussy and regression curve (bold line; Na/Cl ¼ 0.848 0.21 cos t þ 0.259
sin t 0.07 sin 2t þ 0.136 cos 3t þ 0.144 sin 3t þ 0.059 cos 5t 0.06 sin 5t, R2 ¼ 0.50, P < 0.0001).
concentrations exceeded Na concentrations, and Na/Cl was
lowest. Naþ and Cl concentrations were lowest in late
spring and autumn. During summer and autumn, Naþ
concentrations tended to be higher than Cl concentrations, and Na/Cl was highest. The seasonal variation of Cl
concentrations
was
highly
significant
(R2 ¼ 0.21,
P < 0.0001), while the coefficient of determination for Naþ
concentrations was lower but still significant (R2 ¼ 0.12,
P < 0.01). The precipitation volume followed a seasonal
pattern as well (R2 ¼ 0.17, P < 0.001). Naþ and Cl concentrations were highest during the driest period (winter). Naþ
fluxes (R2 ¼ 0.14, P < 0.001) and Cl fluxes (R2 ¼ 0.09,
P < 0.01) tended to be highest in the late spring, early
summer.
3.2. Throughfall
3.2.1. Seasonal variation of the Na/Cl ratio
At Jussy, the seasonal pattern of the Na/Cl ratio in
throughfall samples was much more apparent than in bulk
precipitation, with lower Na/Cl values in the autumn, and
higher values in the spring (Fig. 3). The effects of occasional
contamination with Naþ or Cl prior to 01.01.2001 were
less clear in the throughfall samples than in the bulk
deposition data, because concentrations in throughfall
samples were generally higher. However, only data after
01.01.2001 were included in the regression model with sine
and cosine functions of time, in order to be in a better
position to compare the bulk deposition and throughfall
models. The model resulted in a high coefficient of determination, confirming the strong seasonality of the data
(R2 ¼ 0.50, P < 0.0001).
The site of Jussy is the LWF site where the seasonal
pattern of the Na/Cl ratio was clearest and most systematic.
Jussy is also one of the sites with the highest proportion
(20%) of Na/Cl values outside the range of acceptance. Yet
most of these outlier values can be ascribed to the seasonality of the Na/Cl ratio. When all 11 sites were considered, 15% of all Na/Cl values were outside the range of
acceptance. At the site level, the proportion of values
outside the range varied from 7% (Visp) to 23% (Celerina).
Seasonal patterns were detected on all sites (Table 5). The
seasonality of Na/Cl in throughfall was highly significant in
all broadleaved stands, with peaks either in the spring or in
the summer. The seasonality was less marked at coniferous
sites except at Beatenberg (Picea abies (L.) H.Karst. stand)
and Vordemwald (a mixed stand of Abies alba Mill., Picea
abies and Fagus sylvatica L.).
Table 5
Seasonal variation of Na/Cl in throughfall at 11 LWF sites, sorted according to the vegetation type. The R2 values are the coefficients of determination of the
regression models for the seasonality. The unweighted median Na/Cl ratio, the lower quartile (P25) and the upper quartile (P75) are reported for each site.
Vegetation type
Site
n
Median
P25
P75
R2
P
Peaks
Conifers
Beatenberg
Celerina
National Park
Visp
Vordemwald
Bettlachstock
Lausanne
Othmarsingen
Schänis
Jussy
Novaggio
164
120
162
130
168
123
163
167
161
157
145
0.98
1.20
1.08
0.89
0.74
0.88
0.84
0.90
0.88
0.80
1.02
0.80
0.97
0.89
0.75
0.60
0.73
0.74
0.79
0.73
0.65
0.84
1.21
1.42
1.31
1.06
0.89
1.06
1.03
1.06
1.13
1.00
1.25
0.17
0.07
0.04
0.10
0.10
0.19
0.19
0.32
0.32
0.36
0.20
P < 0.0001
P < 0.05
P < 0.05
P < 0.01
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
Summer
Lowest in spring
Lowest in spring
Spring
Spring
Summer
Summer
Late spring, early summer
Spring
Spring (lowest in autumn)
Summer
Broadleaves
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
Cl- enrichment [meq m-2 (2 weeks)-1]
9112
4
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
3
2
1
0
-1
J A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ
Fig. 4. Temporal variation of the Cl enrichment in throughfall (ECl) at Jussy, calculated as the difference between throughfall and bulk precipitation fluxes
for each sampling period. The data are presented as joined lines without symbols. The regression curve (ECl ¼ 0.25 þ 0.072 cos t 0.16 sin t 0.09
cos 2t 0.15 sin 2t 0.13 cos 3t 0.08 cos 4t þ 0.085 sin 4t; R2 ¼ 0.39, P < 0.0001) is shown as a thicker line.
was stronger than that of Naþ. Peaks of enrichment were
mostly observed either in the spring or in the autumn in
broadleaved stands, and during the growing season in
coniferous stands.
The mean Na/Cl ratio was lower in throughfall than in
bulk precipitation for all sites except the two high elevation
sites (Celerina and National Park) (Tables 3 and 5).
Na+ enrichment [meq m-2 (2 weeks)-1]
3.2.2. Seasonal variation of Naþ and Cl concentrations and
fluxes
At Jussy, throughfall enrichment of Cl showed a sharp
peak in the autumn (Fig. 4). The seasonal model resulted in
a high coefficient of determination R2 of 0.39 (P < 0.0001).
Naþ enrichment in throughfall at Jussy also displayed
a seasonal variation (Fig. 5), with a pronounced peak in the
spring, and a smaller peak in the autumn. Seasonality
explained a smaller fraction of variability in the Naþ
enrichment than in the Cl enrichment (R2 ¼ 0.31 and 0.39
for Naþ and Cl, respectively), but its effect was still highly
significant. Modelling the throughfall fluxes rather than
enrichment resulted in lower coefficients of determination.
The seasonal variation in Naþ and Cl throughfall
enrichment was significant on all sites except Beatenberg
(Picea abies) (Table 6). The seasonality of Cl enrichment
1.6
1.4
1998
1999
2000
2001
3.2.3. Seasonal variation of throughfall fluxes of other nutrients
In order to better understand the processes behind Naþ
and Cl enrichment under forest canopies, we assessed the
throughfall enrichment patterns for the other nutrients
analysed, focusing here again on the sites of Jussy and
Celerina.
At Jussy, throughfall enrichment showed a strong seasonality for all nutrients (Table 7). Two main peaks were
visible for DOC, Kþ and Mg2þ enrichment in the spring and
in the autumn (Fig. 6). There was also a clear enrichment
þ
2þ
peak of NHþ
4 in the spring. Enrichment of DOC, K and Mg
þ
was significantly correlated with enrichment of Na or Cl
(Table 7). Enrichment of inorganic nitrogen (NHþ
4 and NO3 )
tended to be negatively correlated with Cl enrichment,
þ
but there was no correlation between NHþ
4 and Na
2002
2003
2004
2005
2006
2007
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
J A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ A JOJ
Fig. 5. Temporal variation of the Naþ enrichment in throughfall (ENa, see explanation in Fig. 4) at Jussy and regression curve (ENa ¼ 0.11 0.09 cos t þ
0.069 sin t 0.13 sin 2t þ 0.062 sin 3t 0.06 cos 4t; R2 ¼ 0.31, P < 0.0001).
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
9113
Table 6
Seasonal variation of Naþ and Cl throughfall enrichment (ENa and ECl) at 11 LWF sites, sorted according to the vegetation type. The R2 values are the
coefficients of determination of the regression models for the seasonality.
Site
Species
Beatenberg
Picea abies
Celerina
Pinus cembra
National Park
Pinus mugo
Visp
Pinus sylvestris
Vordemwald
Abies alba
Bettlachstock
Fagus sylvatica
Lausanne
Fagus sylvatica
Othmarsingen
Fagus sylvatica
Schänis
Fagus sylvatica
Jussy
Quercus robur
Novaggio
Quercus cerris
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
ENa
ECl
n
R2
P
Peaks
160
160
117
117
160
157
119
118
166
167
125
125
158
159
164
165
161
162
157
158
142
145
0.02
0.06
0.30
0.31
0.11
0.24
0.32
0.28
0.11
0.19
0.09
0.34
0.12
0.14
0.17
0.24
0.07
0.34
0.31
0.39
0.07
0.15
n.s.
P < 0.01
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.001
P < 0.0001
P < 0.01
P < 0.0001
P < 0.001
P < 0.001
P < 0.0001
P < 0.0001
P < 0.01
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.05
P < 0.0001
No clear peaks
Spring (growing season)
Spring (growing season)
Summer (enrichment <0 in winter: road salting?)
Summer (enrichment <0 in winter: road salting?)
Growing season
Growing season
Summer (end of winter)
Summer (end of winter)
Lowest in summer
Autumn
Spring (autumn)
Spring (autumn)
Spring
Autumn (winter)
Spring (autumn)
Autumn
Spring (autumn)
Autumn (spring)
No clear peaks
Autumn
enrichment even though the modelled NHþ
4 enrichment,
like that of Naþ, peaked in the spring.
At Celerina, the seasonal variation of throughfall
enrichment was significant for all elements except Ca2þ
(Fig. 7, Table 8). Peaks of enrichment were less sharp than at
Jussy. Enrichment was positive for Naþ, Cl, Kþ, Mg2þ, DON
and DOC during the growing season. In contrast, enrich
ment of NHþ
4 and NO3 was negative in early summer,
indicating uptake of N by the canopy. Negative net
throughfall for nitrogen has also been reported at other
sites where nitrogen deposition is low (e.g. Lovett and
Lindberg, 1993). Naþ and Cl enrichments tended to peak in
the late spring and were strongly correlated with DOC,
DON, Kþ and Mg2þ enrichments (Table 8).
Table 7
LWF site at Jussy. Coefficients of determination R2 for the seasonal
regression model applied to throughfall enrichment for each nutrient
(data after 01.01.2001), and Spearman coefficients of correlation between
Naþ and Cl enrichment and enrichment for other nutrients (data after
15.05.2001).
Coefficients R2 for seasonal
model
Kþ
NHþ
4
DOC
Mg2þ
Cl
DON
Naþ
NO
3
Ca2þ
2
SO4
Hþ
n
R2
P
196
200
150
198
197
150
196
200
198
200
200
0.46
0.44
0.39
0.38
0.37
0.37
0.29
0.28
0.21
0.14
0
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.05
Spearman coefficients
of correlation r with
Naþ and Cl enrichment
Naþ
Naþ
1
Cl
0.55
Mg2þ 0.49
DOC
0.47
DON
0.47
þ
K
0.40
Ca2þ
0.37
SO2
0.32
4
NHþ
0.13
4
Hþ
0.11
NO3 0.25
P
Cl
P
P < 0.0001 1
P < 0.0001 0.49 P < 0.0001
P < 0.0001 0.48 P < 0.0001
P < 0.0001 0.31 P < 0.001
P < 0.0001 0.45 P < 0.0001
P < 0.0001 0.53 P < 0.0001
P < 0.0001 0.34 P < 0.0001
n.s.
0.27 P < 0.001
n.s.
0.10 n.s.
P < 0.01
0.20 P < 0.05
4. Discussion
4.1. Bulk precipitation
The Na/Cl ratio in bulk precipitation was substantially
higher than the ratio in sea-water at all sites. Similarly,
using data from five European countries, Mosello et al.
(2005) found higher Na/Cl ratios in precipitation samples of
non-marine origin than in samples of marine origin. In
their study, precipitation of non-marine origin was defined
as having an average Cl concentration <50 meq L1, which
was the case with our samples. The Na/Cl ratio in Switzerland is obviously driven by continental sources of Naþ
such as wind-borne soil dust. This dust, as well as other
aerosols or gases containing Naþ or Cl, can be either drydeposited (as our collectors are always open) or washed
from the atmosphere during rain events and thus wetdeposited.
At some of our sites, Na/Cl followed a significant seasonal
variation, with higher ratios in the summer. Shapiro et al.
(2007) also observed a strong seasonality in Na/Cl in wet
precipitation at West Point, New York, approximately
100 km inland from the Atlantic coast. Unlike our findings,
Na/Cl at West Point was lower in the summer and higher in
the winter. They found Na/Cl in the winter was close to the
sea-water ratio, which they ascribed to the influence of large
marine-trajectory storms during colder months. However,
Na/Cl strongly decreased in the summer due to regional
sources of HCl such as coal combustion, waste incineration
or sea-salt dechlorination (a process during which HCl is
produced from the interaction of sea-salt aerosols with
atmospheric acid gases, such as H2SO4 and HNO3). At our
sites, Naþ sources seem to predominate over Cl sources. In
Switzerland, emissions of HCl peaked in the mid 1980s due
to the incineration of increasing volumes of waste (BUWAL,
1995). Since then, the implementation of gas purification
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
Enrichment [g m-2 (2 weeks)-1 or meq m-2 (2 weeks)-1]
9114
7
6
5
4
3
2
1
0
-1
J
Enrichment [dg m-2 (2 weeks)-1 or meq m-2 (2 weeks)-1]
a
Ca2+
ClDOC
K+
Mg+
Na+
F
M
A
M
J
J
A
S
O
N
D
3
b
DON
NH4+
NO3SO42-
2
1
0
-1
J
F
M
A
M
J
J
A
S
O
N
D
Fig. 6. Temporal variation of throughfall enrichment for all elements with a seasonality effect significant at the P < 0.001 level at Jussy. Naþ, Cl, Kþ, Mg2þ, Ca2þ,
2
2
(2 weeks)1, DOC enrichment in g m2 (2 weeks)1, DON enrichment in dg m2 (2 weeks)1.
NHþ
4 , NO3 and SO4 enrichments are in meq m
systems in the waste incineration plants has led to a marked
decrease in HCl emissions, which are expected to soon reach
levels below the levels of the period 1900–1960. At that
time, the main source of HCl was coal combustion for
industry, household use and rail transport.
The seasonal pattern of Na/Cl in bulk precipitation was
most apparent at both sites above 1800 m in the Alps, i.e. in
Celerina and the National Park, and, to a lesser extent, at
mid-elevation sites (Jura, Lower Alps). One process
contributing to the altitude effect is probably the layering of
air masses in winter: clear skies in high altitude regions
contrast with persistent stratus on the Plateau, which acts
like a lid and limits the transport range and deposition of
air-borne pollutants. The strong seasonality observed at
Jussy is in contrast with the other low-altitude sites of this
study. This might be related to the main wind trajectories at
this site (WSW and ENE), which subject Jussy to air masses
originating from the Rhone valley. Emissions from the
chlorine industry in the Grenoble Basin could thus influence the chemistry of the precipitation at Jussy.
4.2. Throughfall
The Na/Cl ratios in throughfall exhibited a stronger
seasonal pattern than in bulk precipitation, especially
under broadleaved canopies. We could show that these
seasonal fluctuations could be related to enhanced
throughfall enrichment of Cl and Naþ, usually in the
Enrichment [dg m-2 (2 weeks)-1 or meq m-2 (2 weeks)-1]
Enrichment [g m-2 (2 weeks)-1 or meq m-2 (2 weeks)-1]
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
0.6
9115
a
ClDOC
K+
Mg+
Na+
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
J
F
M
A
M
J
J
A
S
O
N
D
0.2
b
0.1
0.0
-0.1
-0.2
-0.3
DON
H+
NH4+
NO3-
-0.4
-0.5
-0.6
J
F
M
A
M
J
J
A
S
O
N
D
Fig. 7. Temporal variation of throughfall enrichment for all elements with a seasonality effect significant at the P < 0.001 level at Celerina. Naþ, Cl, Kþ, Mg2þ,
þ
2
(2 weeks)1, DOC enrichment in g m2 (2 weeks)1, DON enrichment in dg m2 (2 weeks)1.
NHþ
4 , NO3 and H enrichments are in meq m
autumn or spring. Canopy exchange (uptake or leaching)
of Cl is usually considered to be negligible in the models
deriving atmospheric deposition from throughfall and
bulk precipitation. In these so-called canopy budget
models, Cl enrichment in throughfall is ascribed to the
dry deposition of HCl in gas form (e.g. Draaijers and
Erisman, 1995). However, the systematically enhanced
enrichment in the autumn that we observed in our
broadleaved stands over several years of measurements is
consistent with the enhanced leaching from senescing
plant tissues reported in previous studies (Tukey, 1970).
Some other in situ studies have also shown that part of
Cl in throughfall can originate from foliage leaching
(Neary and Gizyn, 1994; Houle et al., 1999; Moreno et al.,
2001; Staelens et al., 2007).
The peak of Naþ enrichment in throughfall, which
was more pronounced in the spring e.g. at Jussy, is
consistent with the observation that Naþ can be readily
leached from young leaves (in Tukey, 1970). Other
authors have found enhanced Naþ leaching from
emerging leaves (Staelens et al., 2007). Leaching from
inflorescences could also contribute to enhanced fluxes
in throughfall in May–June.
Patterns of Naþ or Cl enrichments were consistent with
the seasonal patterns for DOC, DON, Kþ and Mg2þ enrichment, which again supports the claim that leaching occurs
from plant tissues (e.g. Parker, 1983; Staelens et al., 2007).
At most of our sites, the seasonality of the Na/Cl ratios can
thus very likely be ascribed to Naþ and Cl leaching from
the canopy, with peaks of leaching occurring at different
9116
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
Table 8
LWF site at Celerina. Coefficients of determination R2 for the seasonal
regression model applied to throughfall enrichment for each nutrient
(data after 01.01.2001), and Spearman coefficients of correlation between
Naþ and Cl enrichment and enrichment for other nutrients (data after
15.05.2001).
Coefficients R2 for seasonal
model
n
NO
3
NHþ
4
þ
147
147
143
K
DOC 115
Mg2þ 143
Naþ 142
DON 114
147
Hþ
141
Cl
2
147
SO4
Ca2þ 143
R2
0.32
0.30
0.26
0.25
0.25
0.24
0.14
0.13
0.11
0.07
0
Spearman coefficients
of correlation r with
Naþ and Cl enrichment
Naþ
P
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.001
P < 0.001
P < 0.01
P < 0.01
P
Cl
p
þ
Na
1
Cl
0.78 P < 0.0001 1
DOC
0.72 P < 0.0001 0.74 P < 0.0001
DON
0.70 P < 0.0001 0.72 P < 0.0001
Mg2þ 0.70 P < 0.0001 0.70 P < 0.0001
Kþ
0.70 P < 0.0001 0.67 P < 0.0001
Ca2þ
0.57 P < 0.0001 0.58 P < 0.0001
þ
H
0.40 P < 0.0001 0.39 P < 0.0001
2
SO4
0.17 n.s.
0.17 n.s.
NO3 0.23 P < 0.05
0.20 P < 0.05
NH4þ 0.39 P < 0.0001 0.42 P < 0.0001
times of year according to the tree species and the length of
the vegetation period, as illustrated by the two examples of
Jussy and Celerina. Additionally, insect attacks might also
influence throughfall enrichment processes. The peaks of
NHþ
4 observed at Jussy in the late spring could be due to
leaching, which Staelens et al. (2007) also observed in
a beech stand during leaf emergence. However, these
peaks, like the phosphorus peaks (data not shown), could
also result from caterpillar outbreaks, which are regularly
observed on this site at this time of year. At Beatenberg,
where we detected no significant seasonal variation in Naþ
or Cl enrichment, the significant seasonality in Na/Cl in
throughfall might simply reflect the seasonality of Na/Cl in
the incident (bulk) precipitation.
Canopy leaching for Naþ and Cl in the spring and
autumn can be estimated, as Staelens et al. (2007) did, by
determining the fraction of throughfall enrichment
corresponding to dry deposition. The latter can be estimated by calculating a so-called dry deposition factor
(DDF), defined as the ratio between throughfall enrichment and bulk precipitation. This DDF is first calculated
on an annual scale, excluding periods (spring and
autumn) when canopy leaching is believed to occur. Dry
deposition of Naþ (or Cl) in the spring or autumn is then
calculated by applying this DDF to the corresponding bulk
deposition. Using this approach at the Jussy site, where
the seasonality of throughfall enrichment was most
marked, we estimated that, on an annual scale, 45% of
Naþ and 35% of Cl in throughfall enrichment originated
from canopy leaching (median value for the period 2002–
2006). The canopy budget models deriving dry deposition
from throughfall measurements should then be modified
to account for the canopy leaching of Naþ and Cl, as e.g.
Staelens et al. (2008) did.
5. Conclusion
In this study, we were able to show that leaching of Naþ
or Cl from the tree canopy occurred at some forest sites in
the autumn or spring. This leaching shows a distinct seasonality for each site and it influences the Na/Cl ratio. The
range of acceptance (0.5–1.5) for this ratio recommended
by the ICP Forests manual to validate chemical analyses
proved useful as, on average, 85% of our Na/Cl values lay
within this range. However, it has to be seasonally adapted
for sites with intense Naþ or Cl canopy leaching in the
spring or autumn. We recommend as a checking procedure
using a plot of the Na/Cl ratio versus time. Furthermore, the
occurrence of Naþ or Cl leaching from the canopy belies
one of the assumptions of the canopy budget models,
which derive dry deposition from throughfall measurements. These models should be adapted to take into
account canopy leaching of these ions.
Acknowledgements
We are grateful to A. Brechbühl, N. Hajjar, O. Schramm,
D. Christen, A. Zürcher and former laboratory staff for
sample handling in the field and in the laboratory, the staff
from the forest services for collecting the samples on the
LWF sites, D. Pezzotta and his team of the WSL central
laboratory for analysing the precipitation samples, P. Jakob
and F. Sutter for the data base support, M. Rebetez for
preparing the wind direction data for Geneva. and S.
Dingwall for editing the English. We gratefully acknowledge the financial support of the Federal Office for the
Environment (FOEN).
References
BUWAL, 1995. Vom Menschen verursachte Luftschadstoff-Emissionen in
der Schweiz von 1900 bis 2010. In: Schriftenreihe Umwelt – Luft, vol.
256. Bundesamt für Umwelt, Wald und Landschaft (BUWAL), Bern,
121 p.
Cherubini, P., Innes, J.L., 2000. Switzerland: the Swiss Long-Term Forest
Ecosystem Research Programme. In: Gosz, J.R., French, C., Sprott, P.,
White, M. (Eds.), The International Long Term Ecological Research
Network. Perspectives from Participating Networks. U.S. Long Term
Ecological Research Network Office, University of New Mexico,
Albuquerque, N.M, pp. 56–59.
de Vries, W., Vel, E., Reinds, G.J., Deelstra, H., Klap, J.M., Leeters, E.E.J.M.,
Hendriks, C.M.A., Kerkvoorden, M., Landmann, G., Herkendell, J.,
Haussmann, T., Erisman, J.W., 2003. Intensive monitoring of forest
ecosystems in Europe: 1. Objectives, set-up and evaluation strategy.
Forest Ecology and Management 174, 77–95.
Draaijers, G.P.J., Erisman, J.W., 1995. A canopy budget model to assess
atmospheric deposition from throughfall measurements. Water, Air,
and Soil Pollution 85, 2253–2258.
FOWG, 2000. Precipitation Maps 1961–1990. Hydrological Atlas of Switzerland. Federal Office for Water and Geology (FOWG), Bern.
Houle, D., Ouimet, R., Paquin, R., Laflamme, J.G., 1999. Interactions of
atmospheric deposition with a mixed hardwood and a coniferous
forest canopy at the Lake Clair Watershed (Duchesnay, Quebec).
Canadian Journal of Forest Research 29, 1944–1957.
Keene, W.C., Pszenny, A.A.P., Galloway, J.N., Hawley, M.E., 1986. Sea-salt
corrections and interpretation of constituent ratios in marine
precipitation. Journal of Geophysical Research Atmospheres 91, 6647–
6658.
Lovett, G.M., Lindberg, S., 1993. Atmospheric deposition and canopy
interactions of nitrogen in forests. Canadian Journal of Forest
Research 23, 1603–1616.
Marchetto, A., Mosello, R., Tartari, G., Derome, J., Derome, K., Sorsa, P.,
König, N., Clarke, N., Ulrich, E., Kowalska, A., 2006. Atmospheric
Deposition and Soil Solution Working Ring Test 2005 – Laboratory
Ring Test for Deposition and Soil Solution Sample Analyses between
the Countries Participating in the ICP Forests Level II Monitoring
Programme. Office National des Forêts, Département Recherche, 85 p.
Moreno, G., Gallardo, J.F., Bussotti, F., 2001. Canopy modification of
atmospheric deposition in oligotrophic Quercus pyrenaica forests of
A. Thimonier et al. / Atmospheric Environment 42 (2008) 9106–9117
an unpolluted region (central-western Spain). Forest Ecology and
Management 149, 47–60.
Mosello, R., Amoriello, M., Amoriello, T., Arisci, S., Carcano, A., Clarke, N.,
Derome, J., Derome, K., Koenig, N., Tartari, G., Ulrich, E., 2005. Validation of chemical analyses of atmospheric deposition in forested
European sites. Journal of Limnology 64, 93–102.
Neary, A.J., Gizyn, W.I., 1994. Throughfall and stemflow chemistry under
deciduous and coniferous forest canopies in south-central Ontario.
Canadian Journal of Forest Research 24, 1089–1100.
Parker, G.G., 1983. Throughfall and stemflow in the forest nutrient cycle.
Advances in Ecological Research 13, 58–135.
Schleppi, P., Waldner, P.A., Stähli, M., 2006. Errors of flux integration
methods for solutes in grab samples of runoff water, as compared to
flow-proportional sampling. Journal of Hydrology 319, 266–281.
Shapiro, J.B., Simpson, H.J., Griffin, K.L., Schuster, W.S.F., 2007. Precipitation chloride at West Point, NY: seasonal patterns and possible
contributions from non-seawater sources. Atmospheric Environment
41, 2240–2254.
Staelens, J., De Schrijver, A., Verheyen, K., 2007. Seasonal variation in
throughfall and stemflow chemistry beneath a European beech
(Fagus sylvatica) tree in relation to canopy phenology. Canadian
Journal of Forest Research 37, 1359–1372.
Staelens, J., Houle, D., De Schrijver, A., Neirynck, J., Verheyen, K., 2008.
Calculating dry deposition and canopy exchange with the canopy
budget model: review of assumptions and application to two deciduous forests. Water, Air, and Soil Pollution 191, 149–169.
9117
Thimonier, A., Schmitt, M., Cherubini, P., Kräuchi, N., 2001. Monitoring the
Swiss forest: building a research platform. In: Anfodillo, T., Carraro, V.
(Eds.), Monitoraggio ambientale: metodologie ed applicazioni. Atti
del XXXVIII Corso di Cultura in Ecologia, 2001. San Vito di Cadore,
Centro Studi per l’Ambiente Alpino, Università degli Studi di Padova,
pp. 121–134.
Thimonier, A., Schmitt, M., Waldner, P., Rihm, B., 2005. Atmospheric
deposition on Swiss Long-term Forest Ecosystem Research (LWF)
plots. Environmental Monitoring and Assessment 104, 81–118.
Tukey Jr., H.B., 1970. The leaching of substances from plants. Annual
Review of Plant Physiology and Plant Molecular Biology 21,
305–324.
Ulrich, E., Mosello, R., Derome, J., Derome, K., Clarke, N., König, N.,
Lövblad, G., Draaijers, G.P.J., 2006. Part VI. Sampling and analysis of
deposition. In: Manual on Methods and Criteria for Harmonized
Sampling, Assessment, Monitoring and Analysis of the Effects of Air
Pollution on Forests. UN-ECE, Convention on Long-Range Transboundary Air Pollution (LRTAP). International Co-operative Programme on Assessment and Monitoring of Air Pollution Effects on
Forests (ICP-Forests), Hamburg Available from: www.icp-forests.org,
74 pp.
Wesely, M.L., Hicks, B.B., 2000. A review of the current status of
knowledge on dry deposition. Atmospheric Environment 34, 2261–
2282.
Witmer, U., 1986. Erfassung, Bearbeitung und Kartierung von Schneedaten in der Schweiz. Geographica Bernensis G25, 215.